I barely was five minutes into prime95 and Worker 4 gave a fatal error about a Hardware Failure Detected. I will attach the results file to see if someone can help.

Well then you have gone to far. Just lay back the frequency and voltage a little and try again. You just need to find that balance of temps/stability/speed.

Those errors are more a symptom of memory problems than CPU. 1.41 volts is plenty high enough for 4.6Ghz in my experience, particularly on a 2500K . Do you have XMP profile set in bios? Try it rather that set memory timings manually. Also consider Vccio in your bios? I dont know what it in called in ASRock(qpi/vtt?) but it can cause similar problems although I would not have expected it at 1333. Edit: Forgot to say that '124' and BSOD are often a symptom of low processor volts. Typically 1.37V or less should be enough for 4.6Ghz.

Those errors are more a symptom of memory problems than CPU. 1.41 volts is plenty high enough for 4.6Ghz in my experience, particularly on a 2500K . Do you have XMP profile set in bios? Try it rather that set memory timings manually. Also consider Vccio in your bios? I dont know what it in called in ASRock(qpi/vtt?) but it can cause similar problems although I would not have expected it at 1333. Edit: Forgot to say that '124' and BSOD are often a symptom of low processor volts. Typically 1.37V or less should be enough for 4.6Ghz.

I agree with the voltage, I know every setup is unique but I am stable on 4.6GHz @ 1.32V offset mode with HT on.

Ive already set the memory timing. That is what I thought was the issue before. Everything with memory is set to: 1600MHz - 9-9-9-25-2T

Try 1.37V for 4.6GHzIf it passes, lower your Vcore and try again till you find the minimumIf it fails, lower your clocks to 4.5GHz. Repeat till you find your maximum overclock / minimum Vcore below 1.37 - 1.38V

Ive already set the memory timing. That is what I thought was the issue before. Everything with memory is set to: 1600MHz - 9-9-9-25-2T

Maybe I am not being clear. Do not set your timings manually. Use the XMP function in bios to load your memory settings:

Prime95 indicated all Cores passed after 1h32min. If they pass can I stop the test? -and I will check the x.m.p mode in bios I believe right now it is auto (or something default) should I change it to profile 1?

What have you changed to pass the test? You can try XMP if you still have problems. You can always change back.

I just brought it down to 4.5 instead of 4.6 @ 1.37 same voltage as 4.6

Here's a faster alternative to prime95 I learned from overclock.net: Open Prime95, select custom. If you have 4GB RAM, then use the standard 1600MB test. If you have 8GB RAM, then type in 6000MB instead. And use 2 minutes per FFT instead of 15 minutes. This way you'll find out in a couple of hours if you are looking good. Then lower Vcore and run the same test again. Do this until you find the lowest possible stable Vcore. After this, run Blend Custom, and again use most of your RAM, but use the standard 15 minutes per FFT. Then let it run for 6 to 12 hrs.

I just brought it down to 4.5 instead of 4.6 @ 1.37 same voltage as 4.6

I don't understand what you're doing Spencer. Did it run Prime95 for 1h32min at 4.6GHz and 1.37V? does that mean that you stopped it or that you got an error after 1h32min?What was your max core temp?What was your Vcore reading at full load in CPU-Z?

No. I ran a test yesterday. That is when I got the error after five minutes. at that time I was running 4.6Ghz @ 1.37v. I stopped the test and today I made a change and ran the same test again. Except this time the only difference was I brought it down to 4.5ghz with the same voltage as yesterday. It did not exceed 1.37v during both tests. Today It passed after an hour in a half. Core Temps did not exceed 68C under full load. Sorry I messed up explaining it earlier. Let me know if you dont understand.

No. I ran a test yesterday. That is when I got the error after five minutes. at that time I was running 4.6Ghz @ 1.37v. I stopped the test and today I made a change and ran the same test again. Except this time the only difference was I brought it down to 4.5ghz with the same voltage as yesterday. It did not exceed 1.37v during both tests. Today It passed after an hour in a half. Core Temps did not exceed 68C under full load. Sorry I messed up explaining it earlier. Let me know if you dont understand.

Ok, so 4.5GHz and 1.37V (in windows, at full load)You can try lowering your Vcore to find the minimum now

Ill start at 1.34v. It just doesnt seem right that you have people running 4.7-4.8ghz @ 1.33-1.34v. But every chip is different.